In a recent paper published in Physical Review Letters (PRL), one of the top physics journals, Jérôme Duplat and Emmanuel Villermaux developed a method to generate centimeter-sized vacuum bubbles in water with miniature laser-driven explosions, and observed the flash of light produced as the bubble collapsed, a not-fully-understood phenomenon known as sonoluminescence. They measured the temperature inside the bubble, spectroscopically, and found it was upwards of 20,000 Kelvin, hotter than any chemical reaction.
Scientific interest in bubbles goes back to Leonardo da Vinci. In the early 20th century the physics of bubbles was investigated more rigorously, because bubbles generated by propellers were causing damage to them. They are also used as contrast agents during medical ultrasound, because they oscillate under an external acoustic field and re-radiate incident sound, which can make it easier to image inside the body. The physics of bubbles is really interesting, but I will only summarize one important tenet: if the pressure inside the bubble is greater than the pressure in the fluid, the bubble grows; if it is lesser, it shrinks.
Sonoluminesence was discovered serendipitously in 1934 during an attempt to use sound waves to speed up photographic development. In 1989 the technology was developed to study it one bubble at a time, by trapping the bubble at the node of an acoustic standing wave. The phenomenon is not fully understood, both experimentally and theoretically. On the experimental side, it has been a matter of debate how hot it actually gets, but if the spectrum is both thermal and visible, it must be in the thousands of degrees. There was a controversial claim that it was hot enough to ignite nuclear fusion, which was met with scepticism. On the theoretical side, it is not known exactly what mechanism produces this light. It may be that the gas inside the bubble is simply heating up as it is compressed, but calculations from the ideal gas law don’t give high enough temperatures. It could be that the rapid acceleration of the wall induces atomic or molecular transitions that give off light, or that the gas becomes ionized into plasma, or it could be bremsstrahlung. The bubbles tend to be so small and so fast that it’s hard to get good spectroscopic information.
The innovation of Duplat and Villermaux was a way to make the bubbles bigger and emptier, by starting them off with a small explosion. Starting with bubbles containing hydrogen and oxygen gas individually, they blast the bubble with a laser to ignite the combustion, driving the expansion of the bubble. The bubble is now full of water vapour, and the molecules get adsorbed into the liquid water at the bubble wall, leaving a mostly-empty centimeter-sized cavity. Now there is a large empty cavity in the water, so the fluid rushes back into it, collapsing it. What vapour that does remain, sonoluminesces. The journal has videos from the experiments, and they are awesome. I particularly recommend video1.avi and MovieS2.mov.
After the laser blast, the flame inside the bubble propagates at 80 meters per second for about 0.1 milliseconds, leaving vapour. The vapour then flies towards the bubble wall at about 1000 meters per second for about 0.01 milliseconds, getting sucked back into its liquid form at the wall. At this point, the pressure inside the bubble is at around a Pascal, (very low, that’s 0.00001 atmospheres), and about 4400 Kelvin. This is where the bubble physics start: it grows at an initial speed of about 10 meters per second for about a millisecond, before collapsing. As it collapses it emits a flash of sonoluminescence, which is recorded through a diffraction grating (to split it into different wavelengths) by a digital camera. Based on the amount of the emitted light, they conclude that the temperature inside the bubble as it collapses is about 26,000 Kelvin.
The novelty of this experiment is not the spectral measurement, but the ability to produce sonoluminescence in such large, empty bubbles. One of the factors that determines the strength of the sonoluminescence effect is the ratio of the bubble’s minimum and maximum size. Not only do they start bigger in this experiment, but they can reach a smaller size because they are so empty, optimizing this ratio. The fact that bigger bubbles take longer to collapse means that it will be easier to watch it happen in real time. In the final paragraph of the paper, they explain that this can be used in the future to better study sonoluminesence.
To summarize, the results of this paper are not novel but the methods are. And I think generating a bubble by igniting a tiny hydrogen fire with a laser pulse is a really awesome thing to do.
*Source of that gif is a movie from the supplemental material of http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.115.094501